Assuming they still work, can I expect a shorter life from them than I would expect from equivalent brand new H9 bulbs?
They'll either >poof< immediately, or last their design life-- the only real degradation they could suffer that impacts life is if the envelope seal is broken.
The softest start in the world is not going to help when the fill gas has been replaced with normal atmospheric gasses. Read again carefully:I wonder if there is a process that could be uncovered, developed by CPF members for CPF members, that could almost eliminate the unpredictable instaflash issue. I know SoftStart drivers help, but, inexplicably, as popular and as desired as SoftStart is, one must dig and beg other members for these now old and used drivers.
Once that seal is broken, even in the tiniest manner, the fill gas (which can be pressurized up to 14ATM) gets out, normal air gets in, and the filament will burn up. Maybe you could bring it up real slow and watch for signs of the filament starting to burn, but properly stored and not dropped or flexed bulbs will be fine after long-term storage.the only real degradation they could suffer that impacts life is if the envelope seal is broken.
The bulbs should be able to withstand below -40F.
headlamp bulbs in Alaska or Nunavut where they instantaneously ignite them several hundreds of times over the period of a winter
Not -40F (or -40C, which is the same temperature). -300F (-196C.) About 77K. 0K is absolute zero, water freezes at 0C (273.15K), boils at 100C (373.15K), so 77K is *COLD*.The bulbs should be able to withstand below -40F...
Could it be because -40C isn't particularly cold as far as a headlamp bulb is concerned?I have not yet heard of a need of special cold resistant headlamp bulbs in Alaska or Nunavut
source: https://ieeexplore.ieee.org/document/4644578scientists said:The behaviour of doped tungsten filaments is discussed with particular reference to the life-controlling processes in incandescent lamps. Since lamp failure is a complex phenomenon depending on many conditions, it appears that there is no single mechanism of universal validity by which the finite life of the tungsten filaments can be explained satisfactorily in all cases. It is suggested that, in addition to the nonuniform evaporation of the filament owing to local defects, other mechanisms, like migration and growth of the potassium-filled bubbles within the wire, and grain-boundary sliding promoted by unfavourable grain shape, may also contribute to the failure of incandescent lamps. Presumably, on the basis of a better understanding of the doping effect and failure mechanisms, further improvements can be achieved in lamp quality by optimisation of bubble dispersion, grain structure and other life-influencing parameters.
There are in fact a couple of issues involved. When cold, the resistance of the filament is shockingly low, if the supply voltage has a low enough impedance, you would probably be shocked at the instantaneous wattage involved (it isn't all that difficult to get to 100kw). At room temperature the resistance is actually quite close to a dead short, you don't need much a positive temperature coefficient when the operating temperature is almost 3000 degrees higher than room temperature to produce substantial resistance at operating temperature!. The other issue is perhaps a bit more interesting. Because of the way filaments are made (they are wound), they actually have a small inductive component,which typically holds down the cold inrush current a little (not much), the bind is that in some circumstances, it has the opposite effect. It depends upon the exact point in the cycle of an AC wave form where the switch is closed. In electronics this problem is often addressed by using an electronic device that only allow the current to start flowing at a zero crossing of the wave form. When I was a University Student, in the lab we had overhead 200 amp circuit breakers whose main function was safety, they were spec'd to open within 8 ms at 110% of rated load. That made them pretty exotic. The bad news was it was surprising just how often they tripped with loads that were under normal circumstances rarely exceeded 5 amps. I happened a lot when the load was a 700 watt Variac transformer.I always assumed the reason for instaflash is a cold filiment self